# Unveiling the Power of Audio-Visual Early Fusion Transformers with Dense Interactions through Masked Modeling

arXiv: https://arxiv.org/abs/2312.01017 · HTML: https://mlsanity.com/paper/2312.01017/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 55 of its 74 references that are in the map.
- First posted on arXiv 2 Dec 2023.

## Builds on (most cited first, 30 of 55)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [SimMIM: A Simple Framework for Masked Image Modeling](https://mlsanity.com/paper/2111.09886/) (2021)
- [Cross-lingual Language Model Pretraining](https://mlsanity.com/paper/1901.07291/) (2019)
- [VGGSound: A Large-scale Audio-Visual Dataset](https://mlsanity.com/paper/2004.14368/) (2020)
- [Masked Feature Prediction for Self-Supervised Visual Pre-Training](https://mlsanity.com/paper/2112.09133/) (2021)
- [Look, Listen and Learn](https://mlsanity.com/paper/1705.08168/) (2017)
- [Masked Autoencoders As Spatiotemporal Learners](https://mlsanity.com/paper/2205.09113/) (2022)
- [Attention Bottlenecks for Multimodal Fusion](https://mlsanity.com/paper/2107.00135/) (2021)
- [Audio-Visual Scene Analysis with Self-Supervised Multisensory Features](https://mlsanity.com/paper/1804.03641/) (2018)
- [ERNIE: Enhanced Representation through Knowledge Integration](https://mlsanity.com/paper/1904.09223/) (2019)
- [Context Autoencoder for Self-Supervised Representation Learning](https://mlsanity.com/paper/2202.03026/) (2022)
- [Masked Autoencoders that Listen](https://mlsanity.com/paper/2207.06405/) (2022)
- [The Sound of Pixels](https://mlsanity.com/paper/1804.03160/) (2018)
- [Cooperative Learning of Audio and Video Models from Self-Supervised Synchronization](https://mlsanity.com/paper/1807.00230/) (2018)
- [SoundNet: Learning Sound Representations from Unlabeled Video](https://mlsanity.com/paper/1610.09001/) (2016)
- [PeCo: Perceptual Codebook for BERT Pre-training of Vision Transformers](https://mlsanity.com/paper/2111.12710/) (2021)
- [Audio-Visual Instance Discrimination with Cross-Modal Agreement](https://mlsanity.com/paper/2004.12943/) (2020)
- [Learning to Localize Sound Source in Visual Scenes](https://mlsanity.com/paper/1803.03849/) (2018)
- [Audio-Visual Segmentation](https://mlsanity.com/paper/2207.05042/) (2022)
- [The Sound of Motions](https://mlsanity.com/paper/1904.05979/) (2019)
- [Unified Multisensory Perception: Weakly-Supervised Audio-Visual Video Parsing](https://mlsanity.com/paper/2007.10558/) (2020)
- [Contrastive Audio-Visual Masked Autoencoder](https://mlsanity.com/paper/2210.07839/) (2022)
- [Should You Mask 15% in Masked Language Modeling?](https://mlsanity.com/paper/2202.08005/) (2022)
- [Music Gesture for Visual Sound Separation](https://mlsanity.com/paper/2004.09476/) (2020)

## Built on by (most cited first, 24 of 24)

- [Text-to-Audio Generation Synchronized with Videos](https://mlsanity.com/paper/2403.07938/) (2024)
- [Multimodal Referring Segmentation: A Survey](https://mlsanity.com/paper/2508.00265/) (2025)
- [MMP: Towards Robust Multi-Modal Learning with Masked Modality Projection](https://mlsanity.com/paper/2410.03010/) (2024)
- [VGGSounder: Audio-Visual Evaluations for Foundation Models](https://mlsanity.com/paper/2508.08237/) (2025)
- [Audio-visual Generalized Zero-shot Learning the Easy Way](https://mlsanity.com/paper/2407.13095/) (2024)
- [DiffGAP: A Lightweight Diffusion Module in Contrastive Space for Bridging Cross-Model Gap](https://mlsanity.com/paper/2503.12131/) (2025)
- [Semantic Grouping Network for Audio Source Separation](https://mlsanity.com/paper/2407.03736/) (2024)
- [COLLAGE: Adaptive Fusion-based Retrieval for Augmented Policy Learning](https://mlsanity.com/paper/2508.01131/) (2025)
- [MLLM4TS: Leveraging Vision and Multimodal Language Models for General Time-Series Analysis](https://mlsanity.com/paper/2510.07513/) (2025)
- [Aligning Audio-Visual Joint Representations with an Agentic Workflow](https://mlsanity.com/paper/2410.23230/) (2024)
- [From Prototypes to General Distributions: An Efficient Curriculum for Masked Image Modeling](https://mlsanity.com/paper/2411.10685/) (2024)
- [Multi-scale Multi-instance Visual Sound Localization and Segmentation](https://mlsanity.com/paper/2409.00486/) (2024)
- [LMAE4Eth: Generalizable and Robust Ethereum Fraud Detection by Exploring Transaction Semantics and Masked Graph Embedding](https://mlsanity.com/paper/2509.03939/) (2025)
- [LSF-Animation: Label-Free Speech-Driven Facial Animation via Implicit Feature Representation](https://mlsanity.com/paper/2510.21864/) (2025)
- [Unified Video-Language Pre-training with Synchronized Audio](https://mlsanity.com/paper/2405.07202/) (2024)
- [Robust Multimodal Learning via Cross-Modal Proxy Tokens](https://mlsanity.com/paper/2501.17823/) (2025)
- [OpenAVS: Training-Free Open-Vocabulary Audio Visual Segmentation with Foundational Models](https://mlsanity.com/paper/2505.01448/) (2025)
- [Towards Relaxed Multimodal Inputs for Gait-based Parkinson's Disease Assessment](https://mlsanity.com/paper/2510.15748/) (2025)
- [GateFusion: Hierarchical Gated Cross-Modal Fusion for Active Speaker Detection](https://mlsanity.com/paper/2512.15707/) (2025)
- [Foley-Flow: Coordinated Video-to-Audio Generation with Masked Audio-Visual Alignment and Dynamic Conditional Flows](https://mlsanity.com/paper/2603.08126/) (2026)
- [If you can describe it, they can see it: Cross-Modal Learning of Visual Concepts from Textual Descriptions](https://mlsanity.com/paper/2411.15611/) (2024)
- [Unmasking Synthetic Realities in Generative AI: A Comprehensive Review of Adversarially Robust Deepfake Detection Systems](https://mlsanity.com/paper/2507.21157/) (2025)
- [A Language-Signal-Vision Multimodal Framework for Multitask Cardiac Analysis](https://mlsanity.com/paper/2508.13072/) (2025)
- [GRAMformer: Any-Order Modality Interactions via Volumetric Multimodal Cross-Attention](https://mlsanity.com/paper/2606.06249/) (2026)

## Abstract (arXiv)

Humans possess a remarkable ability to integrate auditory and visual information, enabling a deeper understanding of the surrounding environment. This early fusion of audio and visual cues, demonstrated through cognitive psychology and neuroscience research, offers promising potential for developing multimodal perception models. However, training early fusion architectures poses significant challenges, as the increased model expressivity requires robust learning frameworks to harness their enhanced capabilities. In this paper, we address this challenge by leveraging the masked reconstruction framework, previously successful in unimodal settings, to train audio-visual encoders with early fusion. Additionally, we propose an attention-based fusion module that captures interactions between local audio and visual representations, enhancing the model's ability to capture fine-grained interactions. While effective, this procedure can become computationally intractable, as the number of local representations increases. Thus, to address the computational complexity, we propose an alternative procedure that factorizes the local representations before representing audio-visual interactions. Extensive evaluations on a variety of datasets demonstrate the superiority of our approach in audio-event classification, visual sound localization, sound separation, and audio-visual segmentation. These contributions enable the efficient training of deeply integrated audio-visual models and significantly advance the usefulness of early fusion architectures.